The Nature of Econometrics and Economic Dataweb.hku.hk/~pingyu/0701/Ch01_Introduction_ver1.pdf ·...

The Nature of Econometrics and Economic Data

Ping Yu

School of Economics and FinanceThe University of Hong Kong

Ping Yu (HKU) Introduction 1 / 34

Course Information

Instructor: Yu, Ping

Email: [email protected]

Teaching Time: Session A: 13:30-14:45pm and 15:00-16:15pm on Thursday;Session B: 9:30-10:45am and 11:00-12:15pm on Friday.

Teaching Location: Session A: CYPP4; Session B: LE4

Office Hour: 3:30-4:30pm, Saturday, KKL1108- I will not answer questions in email if the answer is long or is not easy to explainexactly by words. Please stop by during my office hour.

Tutor: TBA

Email: TBA

Teaching Time: TBA

Teaching Location: TBA

Office Hour: TBA


Information on the Content and Evaluation

Textbook: Introductory Econometrics – A Modern Approach, 6th edition, Jeffery M.Wooldridge [figure here]

Evaluation: HWs (3�5%= 15%), Midterm Test (35%), Final Exam (50%)

HW: HW must be typed. Turn in your HW to the tutor’s assignment box on the dueday. Late HW is not acceptable for whatever reasons. To avoid any risk, start yourHW early.

Tutorial: The answer key to the HWs and midterm would not be posted on moodleand will be discussed by the tutor. The tutorial class starts from week three.Tutorial questions will be posted on Moodle one week in advance. Tutorialquestions are not assignments; there is no need to turn them in.

Examination: Mimic HWs and tutorials. Closed book and closed note. A formulasheet would be provided for the midterm and final and posted on moodle beforethe midterm and final. Slides and Sections indexed by (*) and [Review] are nottested.- Midterm: TBA- Final: TBA


Jeffery M. Wooldridge (1960-),MSU, 1986UCSDPhD


Course Policy

In Class: (i) turn off your cell phone and keep quiet; (ii) come to class and returnfrom the break on time; (iii) you can ask me freely in class, but if your question isfar out of the course or will take a long time to answer, I will answer you after class.

Plagiarism: If judged as “plagiarism”, you are in serious trouble.

Policy: If a few students are judged to copy each other, each gets zero mark. I willnot judge who copied whom. So DO NOT copy others and DO NOT be copied byothers.- You may discuss with your classmates about HW, but DO NOT copy each other.- This policy applies to HW, midterm and final.

Feedback: Any feedback to my teaching (e.g., the lecturer’s English is hard tofollow, technicalities are too hard to understand, the teaching should slow down,more interactions are required, etc.) is very welcome. I would incorporate yourfeedbacks in my future teaching during the semester. You can also give yourfeedbacks to the TA so that the TA can discuss them in tutorial classes.


Course Outline

Week 1: Chapter 1 - Introduction

Week 2, 3, 4: Chapter 2 - The Simple Regression Model

Week 5, 6: Chapter 3 - Multiple Regression Analysis: Estimation

Week 7, 8: Chapter 4 - Multiple Regression Analysis: Inference

Week 9: Chapter 6 - Multiple Regression Analysis: Further Issues

Week 10: Chapter 7 - Regression Analysis with Qualitative Information

Week 11: Chapter 8 - Heteroskedasticity

Week 12: Chapter 9 - More on Specification and Data Issues

Appendix B and C (Review of Probability and Statistics) are discussed only ifrequired.

I will roughly follow the textbook.


What is Econometrics?




What Is Econometrics?

Econometrics = use of statistical methods to analyze economic data.

Econometricians typically analyze nonexperimental data (or observationaldata/retrospective data, to emphasize that the researcher is a passive collector ofthe data).

Typical goals of econometric analysis:- Estimating relationships between economic variables;- Testing economic theories and hypotheses;- Forecasting economic variables;- Evaluating and implementing government and business policy.


Steps in Empirical Economic Analysis





An empirical analysis uses data to estimate a relationship or to test a theory,which are important in forecasting and policy analysis.

Step 1: Economic model (this step is often skipped)- Maybe micro- or macro- models- Often use optimizing behavior, equilibrium modeling, � � �- Establish relationships between economic variables- Example: utility maximization =) demand equations:

D = f (prices, income, taste) ,

where prices include the price of the good, and the prices of substitute andcomplementary goods.

Step 2: Econometric model

We will provide two examples for these two steps.



Economic Model of Crime

Derives equation for criminal activity based on utility maximization:

y = f (x1,x2,x3,x4,x5,x6,x7),

wherey = hours spent in criminal activitiesx1 = "wage" for an hour spent in criminal activityx2 = hourly wage for legal employmentx3 = income other than from crime or employmentx4 = probability of getting caughtx5 = probability of being convicted if caughtx6 = expected sentence if convictedx7 = age

Functional form of relationship is not specified because it depends on anunderlying utility function which is rarely known.

Equation could have been postulated without economic modeling.



History of Crime Analysis

Becker, G.S., 1968, Crime and Punishment: An Economic Approach, Journal ofPolitical Economy, 76, 169-217.

Gary S. Becker (1930-2014), Chicago, 1992NP, 1955ChicagoPhD



Job Training and Worker Productivity

What is effect of additional training on worker productivity?

Formal economic theory not really needed to derive equation:

wage = f (educ,exper , training)

wherewage = hourly wageeduc = years of formal educationexper = years of workforce experiencetraining = weeks spent in job training

Other factors, e.g., productivity, may be relevant, but these are the most important.- maybe productivity can be mostly explained by these three factors.



Econometric Model of Criminal Activity

The functional form has to be specified.

Variables may have to be approximated by other quantities:

crime = β 0+β 1wagem+β 2othinc+β 3freqarr +β 4freqconv

+β 5avgsen+β 6age+u,

wherecrime = some measure of the frequency of criminal activitywagem = the wage that can be earned in legal employmentothinc = the income from other sources (assets, inheritance, and so on)freqarr = the frequency of arrests for prior infractions (to approximate the

probability of arrest)freqconv = the frequency of convictionavgsen = the average sentence length after conviction

u is unobserved determinants of criminal activity, e.g., moral character, wage incriminal activity, family background,� � �



Econometric Model of Job Training and Worker Productivity

An econometric model of job training might be

wage = β 0+β 1educ+β 2exper +β 3training+u,

where educ, exper and training are defined above, and u is unobserveddeterminants of the wage, e.g., innate ability, quality of education, familybackground,� � �Most of econometrics deals with the specification of the error u.

Econometric models may be used for hypothesis testing.- E.g., the parameter β 3 represents effect of training on wage.- How large is this effect? Is it different from zero?


The Structure of Economic Data





Econometric analysis requires data.

Different kinds of economic data sets:- Cross-sectional data- Time series data- Pooled cross sections- Panel/Longitudinal data

Econometric methods depend on the nature of the data used.- Use of inappropriate methods may lead to misleading results.



a: Cross-Sectional Data

Sample of individuals, households, firms, cities, states, countries, or other units ofinterest at a given point of time/in a given period.

Cross-sectional observations are more or less independent.- E.g., random sampling from a population.

Ordering of observations is not important.

Sometimes "pure" random sampling is violated, e.g., units refuse to respond insurveys, or if sampling is characterized by clustering.

Typical applications: applied microeconomics.



Cross-Sectional Data on Wages and Other Characteristics



Cross-Sectional Data on Growth Rates and Country Characteristics



b: Time Series Data

Time series data consist observations of a variable or several variables over time.- E.g., stock prices, money supply, consumer price index, gross domestic product,annual homicide rates, automobile sales,� � �Time series observations are typically serially correlated.

Ordering of observations conveys important information.

Data frequency: daily, weekly, monthly, quarterly, annually,� � �Typical features: trends and seasonality.

Typical applications: applied macroeconomics and finance.



Time Series Data on Minimum Wages and Related Variables



c: Pooled Cross Sections

Two or more cross sections are combined in one data set.

Cross sections are drawn independently of each other.

Pooled cross sections often used to evaluate policy changes.

Example:- Evaluate effect of change in property taxes on house prices.- Random sample of house prices for the year 1993.- A new random sample of house prices for the year 1995.- Compare before/after (1993: before reform, 1995: after reform).



Pooled Cross Sections on Housing Prices



d: Panel or Longitudinal Data

The same cross-sectional units are followed over time.

Panel data have a cross-sectional and a time series dimension.

Panel data can be used to account for time-invariant unobservables.

Panel data can be used to model lagged responses.

Example:- City crime statistics; each city is observed in two years.- Time-invariant unobserved city characteristics may be modeled.- Effect of police on crime rates may exhibit time lag.



Two-Year Panel Data on City Crime Statistics


Causality and the Notion of Ceteris Paribus in Econometric Analysis

Causality and the Notion of Ceteris Paribusin Econometric Analysis




Definition of causal effect of x on y :

How does variable y change if variable x is changed

but all other relevant factors are held constant.

Most economic questions are ceteris paribus questions.

It is important to define which causal effect one is interested in.

It is useful to describe how an experiment would have to be designed to infer thecausal effect in question.

Let’s check four examples below; the first three deal with cross-sectional data atvarious levels of aggregation (e.g., at the individual or city levels), and the last onedeals with time series data.

The key point for all examples is that we need to randomly assign x such that allother relevant factors are balanced. (see Chapter 2 for a rigorous treatment).



Effects of Fertilizer on Crop Yield

The question: By how much will the production of soybeans increase if oneincreases the amount of fertilizer applied to the ground?

Implicit assumption: all other factors that influence crop yield such as quality ofland, rainfall, presence of parasites etc. are held fixed.

Experiment:- choose several one-acre plots of land;- randomly assign different amounts of fertilizer to the different plots;- compare yields.

Experiment works because amount of fertilizer applied is unrelated to other factorsinfluencing crop yields.



History of Production Function Estimation

Griliches, Z., 1957, Specification Bias in Estimates of Production Functions,Journal of Farm Economics, 39, 8-20.

Zvi Griliches (1930-1999), Harvard, 1957ChicagoPhD



Measuring the Return to Education

The question: If a person is chosen from the population and given another year ofeducation, by how much will his or her wage increase?

Implicit assumption: all other factors that influence wages such as experience,family background, intelligence etc. are held fixed.

Experiment:- choose a group of people;- randomly assign different amounts of education to them (infeasible!);- compare wage outcomes.

Problem without random assignment: amount of education is related to otherfactors that influence wages (e.g. intelligence).



Effect of Law Enforcement on City Crime Levels

The question: If a city is randomly chosen and given ten additional police officers,by how much would its crime rate fall?- Alternatively: If two cities are the same in all respects, except that city A has tenmore police officers, by how much would the two cities crime rates differ?

Experiment:- randomly assign number of police officers to a large number of cities;- compare crime rates.

In reality, number of police officers will be determined by crime rate (simultaneousdetermination of crime and number of police).



Effect of the Minimum Wage on Unemployment

The question: By how much (if at all) will unemployment increase if the minimumwage is increased by a certain amount (holding other things fixed)?

Experiment:- Government randomly chooses minimum wage each year and observesunemployment outcomes;

Experiment will work because level of minimum wage is unrelated to other factorsdetermining unemployment;

In reality, the level of the minimum wage will depend on political and economicfactors that also influence unemployment.



Testing Predictions of Economic Theories

Economic theories are not always stated in terms of causal effects, but they oftenhave predictions that can be tested using econometric methods.

For example, the expectations hypothesis states that long term interest rates equalcompounded expected short term interest rates,

(1+ rlt )n =

�1+ re

year1

��1+ re

year2

��

�1+ re

yearn�,

where rlt is the long term interest rate, and reyear i is the expected interest rate of

year i .

An implication is that the interest rate of a three-months T-bill should be equal tothe expected interest rate for the first three months of a six-months T-bill;otherwise, there would be arbitrage (why?).

This implication can be tested using econometric methods.


The Nature of Econometrics and Economic Dataweb.hku.hk/~pingyu/0701/Ch01_Introduction_ver1.pdf ·...

Documents

Transcript of The Nature of Econometrics and Economic Dataweb.hku.hk/~pingyu/0701/Ch01_Introduction_ver1.pdf ·...